Bt refuge theory unravels

May 2013
Canola field in Washington County. Photo Gary Halvorson,
Oregon State Archives [Attribution], via Wikimedia Commons
US regulators and the biotech industry realised from day one that, if farmers grew monocultures of GM crops which produce a single insecticide, it was inevitable that pests resistant to the novel toxin would emerge.

To delay this inconvenience, they devised a 'refuge' strategy. Farmers are required to plant areas of conventional plants to harbour a population of susceptible pests. The theory is that two resistant mutant insects must breed together to produce resistant offspring so that breeding with the normal insects from the refuge will dilute the chances of this happening.

Since the refuge strategy will delay, but not prevent, the spread of resistant pests, the biotech industry has produced a new generation of GM crops designed to be resistance-proof. These latest GM offerings are 'pyramided' with more than one 'Bt' insecticidal toxin variant. In theory, if one type of Bt protein doesn't kill the pest, the other one will, and the possibility of an insect developing resistance to both at once is highly remote.

All these theories played out well in simulation models.

Before the first-generation, single-toxin, Bt crops hit the fields, scientists were recommending a non-GM refuge area of 50% of the biotech crop. This suggestion wasn't very attractive or profitable, and it immediately got watered down by regulators under pressure from industry to 20%.

With the advent of pyramided multi-toxin Bt seed, the industry successfully argued that refuges as low as 5-10% would be adequate. At this level, it’s possible to market 'refuge-in-a-bag' seed blends of GM/non-GM seed.

And so now, the regulators are happy they're doing a fine job of regulating GM, and industry is happy it has secured the entire seed market and can be confident its profits are safe.

But are the farmers happy? Are the pests dead in the field? Or, are things out there in the real world a bit more complicated?

Early on, it became apparent that insect-resistance to single Bt-toxins in GM crops was evolving much faster than the computer had predicted. Within 15 years, several pests had developed resistance. One reason may have been that the mandatory 20% refuge is a maximum: who knows how much its effectiveness has been reduced by excessive distance from the GM crop, by insecticide applications which kill off the necessary non-resistant breeding stock, or simply by farmer non-compliance.

Also complicating the picture is the bit the biotech industry won't admit: GM is not a precise nor predictable technology. The actual amounts of the toxic Bt protein produced in different parts of the plant at different times of year and under different weather conditions, have been found to be highly variable. If the Bt protein is too dilute to kill the pests, it creates a perfect scenario in which resistance will evolve. The assumed need for both parents to be resistant in order to produce resistant offspring (the central plank of the refuge strategy) has also proven false: one mutant parent pest is sometimes all it takes for the old scourge to make a comeback.

However, now that the pyramided double-toxin Bt strains are in vogue and the 'refuges' have all but disappeared, will biotech second-generation GM crop-protection finally work as intended?

Recent research by the University of Arizona suggests not.

Contrary to the central theory that different artificial Bt proteins will require the evolution of different mechanisms to confer resistance, cross-resistance between commonly used Bt toxins has been observed in 19 out of 21 experiments on the subject. This problem will likely be exacerbated by the historical presence of single toxins and by the decline in levels of both toxins during the growing season.

One of the authors commented that regulators have made overly optimistic assumptions in reducing the requirement for refuges. He said “simulations tell us that with 10 percent of acreage set aside for refuges, resistance evolves quite fast, but if you put 30 or 40 percent aside, you can substantially delay it.”


If the refuges are going to have any effect, science seems to indicate that they must be substantial, 30-50% of the size of the GM crop depending on how many Bt toxins are present.

Because of the erosion of biotech industry profit this would entail, it isn't ever going to happen. Instead, we'll see a tread-mill of multiple Bt and other fancier artificial insecticidal proteins spreading through our food chain and environment.

And are they safe?

The testing of single Bt proteins has been minimal, short-term and is based on an analogue of the protein not on the actual substance produced by the GM plant and consumed. Pyramided Bt toxins haven't been tested at all.

If their effects can cross-react so unexpectedly in insects, why not inside you too?

  • Thierry Brévault et al., 2013, Potential shortfall of pyramided transgenic cotton for insect resistance management, PNAS
  • Daniel Stolte, Multi-toxin biotech crops not silver bullets, UA scientists warn, University of Arizona News, 28.03.13
  • Susan Jongeneel, Expect more soil insecticide used with Bt hybrids, AG Professional, 1.04.13

No comments:

Post a Comment

Thanks for your comment. All comments are moderated before they are published.